5.1 Differential expression of CTHRC1 in non-tumor and tumor tissues
In a first step, significant difference was observed for expression of CTHRC1 mRNA in non-tumor and tumor tissues. By analyzing the gene chip data of Oncomine, expression levels of CTHRC1 are high in most tumor tissues, consisting of brain and CNS cancer, breast cancer, cervical cancer, colorectal cancer, esophageal cancer, gastric cancer, head and neck cancer, kidney cancer, leukemia, liver cancer, lung cancer, lymphoma, melanoma, ovarian cancer, pancreatic cancer and prostate cancer. In TIMER database, Compared with the normal counterparts, the transcript level of CTHRC1 was over-expressed in most cancer types, including BLCA (bladder urothelial carcinoma), BRCA (breast invasive carcinoma), CHOL (cholangio carcinoma), COAD (colon adenocarcinoma), HNSC (head and neck cancer), ESCA (esophageal carcinoma), KIRC (kidney renal clear cell carcinoma), KIRP (kidney renal papillary carcinoma), LIHC (liver hepatocellular carcinoma), LUAD (lung adenocarcinoma), LUSC (lung squamous cell carcinoma), PRAD (prostate adenocarcinoma), READ (rectum adenocarcinoma), STAD (stomach adenocarcinoma), THCA (thyroid carcinoma) and UCEC(uterine corpus endometrial carcinoma). Most of the results of the UALCAN analysis bear a close resemblance data from the TIMER database. And CTHRC1 mRNA was also upregulated in CESC (cervical squamous cell carcinoma and endocervical adenocarcinoma), GBM (glioblastoma multiforme), KICH (kidney chromophobe), SARC (sarcoma) and THYM (thymoma). In TNM-plot, CTHRC1 mRNA was also upregulated in BLCA, BRCA, COAD, ESCA, LIHC, LUAD, LUSC, OV, PAAD (pancreatic adenocarcinoma), PRAD, READ, KIRC, KIRP, SKCM (skin cutaneous melanoma), STAD, TGCT (testicular germ cell tumors), UCS (uterine carcinosarcoma), ACC (adrenocortical carcinoma), LAML (acute myeloid leukemia) and UCEC. The above results tentatively proposed that CTHRC1 was an oncogene and had potential as a diagnostic marker for pan-cancer.
5.2 CTHRC1 and prognosis of cancer patients
Our study next found that over-expression of CTHRC1 affects a poor prognosis of patients based on GEPIA2.0 (TCGA RNA-seq data) and KM-plotter (gene chip data). Detailed results are presented in the following sets of data. In GEPIA2.0, ACC [OS: HR= 4.2, P= 0.00039, DFS: HR= 5, P <5.6e-06]; COAD [OS: HR= 1.8, P= 0.018, DFS: HR =1.8, P =0.015]; KIRP [OS: HR= 2, P= 0.029, DFS: HR= 1.9, P= 0.027]; BLCA [OS: HR= 1.4, P= 0.045]; LGG [OS: HR= 1.6, P= 0.015]; LIHC [OS: HR= 1.6, P= 0.0094]; SARC [OS: HR= 1.8, P= 0.0059]; STAD [OS: HR =1.4, P =0.049]; KICH [DFS: HR =4.4, P =0.04]; KIRC [DFS: HR =2.2, P <2.1e-05]; PRAD [DFS: HR =1.8, P= 0.0064]. In KM-plotter, OV [OS: HR= 1.41, P= 0.000087, PFS: HR =1.41, P= 0.00028, PPS: HR= 1.3, P= 0.032]; BRCA [DMFS: HR= 1.32, P= 0.025]. This result implies that CTHRC1 has the potential to be used as a prognostic marker in pan-cancer.
5.3 Relative expression levels of CTHRC1 in different tumor subtypes (immune and molecular subtypes)
Using the Subtype module of the TISIDB database, we found that CTHRC1 transcript levels significantly varied in immune subtypes [ C1 (wound healing), C2 (IFN-gamma dominant), C3 (inflammatory), C4 (lymphocyte depleted), C5 (immunologically quiet) and C6 (TGF-b dominant)] of multiple tumors, which include BLCA, BRCA, COAD, KIRP, KICH, KIRC, LIHC, SARC, PRAD, OV and LGG. CTHRC1 was not always highly expressed in a single immune subtype. In more detail, CTHRC1 showed the highest expression in C1 subtype of three distinct tumors, which include KIRC, LIHC and SARC. Highest expression in C2 subtype, which include BLCA, KIRP and PRAD. Highest expression in C3 subtype, which include KICH, OV and LGG. Highest expression in C4 subtype, which include BRCA and COAD. Furthermore, CTHRC1 expression is also differentially expressed in molecular subtypes of multiple tumors, which include BLCA, BRCA, COAD, KIRP, KICH, KIRC, LIHC, SARC, PRAD, OV and LGG. The above analysis indicated that CTHRC1 has a diagnostic index as a classification of pan-cancer.
5.4 CTHRC1 CNV and Methylation in tumors
The use of cBioPortal database was used to analyse the uniform sample (TCGA, PanCancer Atlas). CTHRC1 amplification (AMP) was the most prevalent actionable alteration in multiple tumors, consisting of OV, BLCA, BRCA, LIHC, PRAD, STAD, SARC, COAD, KICH, LGG and KIRC. And CTHRC1 CNV alteration was a high risk factor in cancer patients, consisting of OV, LGG, SARC, PRAD and KIRP. For instance, the group of CTHRC1 CNV alteration had a significantly poorer OS, PFS and DSS. More detailed data are shown in (Figure 4). DNA methylation was considered an essential factor impacting on the occurrence and progression of tumor. Currently, there is a paucity of research on the relationship between abnormal methylation of CTHRC1 gene and tumors. Thus, through the analysis results of the SMART database, we demonstrated that abnormal methylation of CTHRC1 gene in a variety of tumor tissues, including BLCA, BRCA, CESC, COAD, ESCA, HNSC, KIRC, LIHC, LUAD, LUSC, READ, THCA and UCEC. Also, we found that CTHRC1 hypermethylation was generally linked to good prognosis in BLCA, BRCA, HNSC, KIRC and UCEC. More detailed results are shown in (Figure 5).
5.5 CTHRC1 and human cancers immune microenvironment
The TIMER database found a link between CTHRC1 transcript levels and immune cell numbers in a variety of cancer types. For example, a significant positive association was observed between CTHRC1 and multiple immune cell types such as B cells, CD4 +T cells CD8 +T cells, neutrophils, macrophages and dendritic cells in BRCA. An additional thing we noticed was that CTHRC1 was not always related to specific immune cell species. Like, CTHRC1 has the largest correlation coefficient with B cells in ten cancer types. Having the closest association with CD4+T cells in eight cancer types. Having the largest correlation coefficient with CD8+T cells in ten cancer types. Having the closest association with macrophages in ten cancer types. Having the closest association with neutrophils in ten cancer types. And having the closest association with dendritic cells in twelve cancer types. This suggests that CTHRC1 exerts different immune effects in different types of tumor tissues. The sangerbox platform analyzed 47 immune checkpoint genes and found a positive correlation between CTHRC1 and immune checkpoint genes in most tumor tissues. In addition, there was a significant positive correlation between CTHRC1 and CD276 in most tumor tissues. TMB, MSI and neoantigens were important evaluation metric of immunotherapy efficacy. There was a positive correlation between CTHRC1 with TMB in COAD, LAML, LGG, LUAD, PRAD, SARC and THYM, and a negative correlation between CTHRC1 with TMB in GBM, LIHC and UCEC. MSI and neoantigens exhibited the closest linkage with CTHRC1. In particular, UCEC patients with high CTHRC1 expression tend to show negative scores of TMB, MSI and neoantigens. This suggests that UCEC patients with high CTHRC1 expression have poor immune efficacy.